• Journal of Korean Society of Industrial and Systems Engineering
• /
• v.32 no.2
• /
• pp.132-139
• /
• 2009

Yield is a very important measure that can expresses simply for productivity and performance of company. So, yield is used widely in many industries nowadays. With the development of the information technology and online based real-time process monitoring technology, many industries operate the production lines that are developed into automation system. In these production lines, the product structures are very complexity and variety. So, there are many multi-variate processes that need to be monitored with many quality characteristics and associated process variables at the same time. These situations have made it possible to obtain super-large manufacturing process data sets. However, there are many difficulties with finding the cause of process variation or useful information in the high capacity database. In order to solve this problem, neural networks technique is a favorite technique that predicts the yield of process for process control. This paper uses a neural networks technique for improvement and maintenance of yield in manufacturing process. The purpose of this paper is to model the prediction of a sub process that has much effect to improve yields in total manufacturing process and the prediction of adjustment values of this sub process. These informations feedback into the process and the process is adjusted. Also, we show that the proposed model is useful to the manufacturing process through the case study.

• Journal of Information Technology Applications and Management
• /
• v.13 no.4
• /
• pp.273-290
• /
• 2006

This study suggests a frequency matrix technique to predict personal credit rate more efficiently using incomplete data sets. At first this study test on multiple discriminant analysis and logistic regression analysis for predicting personal credit rate with incomplete data sets. Missing values are predicted with mean imputation method and regression imputation method here. An artificial neural network and frequency matrix technique are also tested on their performance in predicting personal credit rating. A data set of 8,234 customers in 2004 on personal credit information of Bank A are collected for the test. The performance of frequency matrix technique is compared with that of other methods. The results from the experiments show that the performance of frequency matrix technique is superior to that of all other models such as MDA-mean, Logit-mean, MDA-regression, Logit-regression, and artificial neural networks.

• Proceedings of the Korean Operations and Management Science Society Conference
• /
• 1995.09a
• /
• pp.334-343
• /
• 1995

In recent years, neural networks have been developed as an alternative to traditional statistical techniques. In this study, a neural network model was compared to traditional forecasting models in terms of their capabilities to forecast passenger traffic for flights between U.S. and Korea. The results show that the forecasting ability of the neural networks was superior to the traditional models. In terms of accuracy, the performance of the neural networks was quite encouraging. Using mean absolute deviation, the neural network performed best. The new technique is easy to learn and apply with commercial neural network software. Therefore, airline decision makers should benefit from using neural networks in forecasting passenger loads.

• Proceedings of the KIEE Conference
• /
• 1995.07b
• /
• pp.971-973
• /
• 1995

This paper presents a new design technique that can be used for brain-state-in-a-box neural networks to realize associative memories. The applicability of the technique is demonstrated by means of a simulation example, which illustrates its strengths.

• Journal of Intelligence and Information Systems
• /
• v.5 no.1
• /
• pp.95-101
• /
• 1999

In this paper, a comparison between different methods to combine predictions from neural networks will be given. These methods are bagging, bumping, and balancing. Those are based on the analysis of the ensemble generalization error into an ambiguity term and a term incorporating generalization performances of individual networks. Neural Networks and AI machine learning models are prone to overfitting. A strategy to prevent a neural network from overfitting, is to stop training in early stage of the learning process. The complete data set is spilt up into a training set and a validation set. Training is stopped when the error on the validation set starts increasing. The stability of the networks is highly dependent on the division in training and validation set, and also on the random initial weights and the chosen minimization procedure. This causes early stopped networks to be rather unstable: a small change in the data or different initial conditions can produce large changes in the prediction. Therefore, it is advisable to apply the same procedure several times starting from different initial weights. This technique is often referred to as training ensembles of neural networks. In this paper, we presented a comparison of three statistical methods to prevent overfitting of neural network.

• Journal of applied mathematics & informatics
• /
• v.28 no.1_2
• /
• pp.295-309
• /
• 2010

In this paper, we study delayed high-order Cohen-Grossberg neural networks with reaction-diffusion terms and Neumann boundary conditions. By using inequality techniques and constructing Lyapunov functional method, some sufficient conditions are given to ensure the existence and convergence of the periodic oscillatory solution. Finally, an example is given to verify the theoretical analysis.

• Journal of the Korean Mathematical Society
• /
• v.48 no.2
• /
• pp.223-240
• /
• 2011

In this paper, the problem on periodic solutions of the bidirectional associative memory neural networks with both periodic coefficients and periodic time-varying delays is discussed. By using degree theory, inequality technique and Lyapunov functional, we establish the existence, uniqueness, and global asymptotic stability of a periodic solution. The obtained results of stability are less restrictive than previously known criteria, and the hypotheses for the boundedness and monotonicity on the activation functions are removed.

• Journal of the Korean Mathematical Society
• /
• v.43 no.2
• /
• pp.445-459
• /
• 2006

In this paper cellular neural networks with continuously distributed delays are considered. Sufficient conditions for the existence and exponential stability of the almost periodic solutions are established by using fixed point theorem, Lyapunov functional method and differential inequality technique. The results of this paper are new and they complement previously known results.

• Journal of the Institute of Electronics Engineers of Korea SP
• /
• v.42 no.4 s.304
• /
• pp.39-48
• /
• 2005

This paper proposes novel texture segmentation method using Bayesian estimation method and neural networks. We use multi-scale wavelet coefficients and the context information of neighboring wavelets coefficients as the input of networks. The output of neural networks is modeled as a posterior probability. The context information is obtained by HMT(Hidden Markov Tree) model. This proposed segmentation method shows better performance than ML(Maximum Likelihood) segmentation using HMT model. And post-processed texture segmentation results as using multi-scale Bayesian image segmentation technique called HMTseg in each segmentation by HMT and the proposed method also show that the proposed method is superior to the method using HMT.

• Korean Management Science Review
• /
• v.11 no.2
• /
• pp.133-146
• /
• 1994

This study compares the performance of neural networks and ordinary least squares regression with quality-control processes. We examine the applicability of neural networks because they do not require any assumptions regarding either the functional from of the underlying process or the distribution of errors. The coefficient of determination($R^2$), mean absolute deviation(MAD), and the mean squared error(MSE) metrics indicate that neural networks are a viable and can be a superior technique. We also demonstrate that an assessment of the magnitude of the neural notwork input layer cumulative weights can be used to determine the relative importance of predictor variables.